This paper introduces a fully deep learning approach to monocular SLAM, which can perform simultaneous localization using a neural network for learning visual odometry (L-VO) and dense 3D mapping. Dense 2D flow and a depth image are generated from monocular images by sub-networks, which are then used by a 3D flow associated layer in the L-VO network to generate dense 3D flow. Given this 3D flow...

The submodules with the property of the title ( a submodule $N$ of an $R$-module $M$ is called strongly dense in $M$, denoted by $Nleq_{sd}M$, if for any index set $I$, $prod _{I}Nleq_{d}prod _{I}M$) are introduced and fully investigated. It is shown that for each submodule $N$ of $M$ there exists the smallest subset $D'subseteq M$ such that $N+D'$ is a strongly dense submodule of $M$ and $D'bi...

Most texture-based 3D flow visualization techniques such as volume Line Integral Convolution (LIC) are limited to steady fields. Time-varying dense volume flow visualization is still an open problem due to intensive computation, demanding temporal-spatial coherence, and ad-hoc volume rendering. Our work is based on an enhanced Unsteady Flow LIC (UFLIC) algorithm, Accelerated UFLIC (AUFLIC), whi...

It is several decades that the application of membrane processes for gas separation has become a great concern within industries. Among them, the polymeric membranes of different structures have played the most important role. Such application has been widely extended to gas separation and due to the necessity of both; prediction of the behavior of these processes and recognition of the effecti...